Genome mapping of postzygotic hybrid necrosis in an interspecific pear population

Abstract

Deleterious epistatic interactions in plant inter- and intraspecific hybrids can cause a phenomenon known as hybrid necrosis, characterized by a typical seedling phenotype whose main distinguishing features are dwarfism, tissue necrosis and in some cases lethality. Identification of the chromosome regions associated with this type of incompatibility is important not only to increase our understanding of the evolutionary diversification that led to speciation but also for breeding purposes. Development of molecular markers linked to the lethal genes will allow breeders to avoid incompatible inbred combinations that could affect the expression of important agronomic tratis co-segregating with these genes. Although hybrid necrosis has been reported in several plant taxa, including Rosaceae species, this phenomenon has not been described previously in pear. In the interspecific pear population resulting from a cross between PEAR3 (Pyrus bretschneideri × Pyrus communis) and 'Moonglow' (P. communis), we observed two types of hybrid necrosis, expressed at different stages of plant development. Using a combination of previously mapped and newly developed genetic markers, we identified three chromosome regions associated with these two types of lethality, which were genetically independent. One type resulted from a negative epistatic interaction between a locus on linkage group 5 (LG5) of PEAR3 and a locus on LG1 of 'Moonglow', while the second type was due to a gene that maps to LG2 of PEAR3 and which either acts alone or more probably interacts with another gene of unknown location inherited from 'Moonglow'.

Figure 1

HN phenotypes in the Pyrus interspecific PEAR3 × ‘Moonglow’ population. Three distinct phenotypes were observed in the seedlings. Pictures were taken 30 days after germination: (a) ‘Type 1’ seedlings had stopped growing and chlorosis and necrotic lesions were apparent on their leaves; (b) ‘Type 2’ seedlings initially grew normally; however, their leaves began to cup downwards and to become chlorotic and necrotic. (c) ‘Type 3’ seedlings grew normally.

Figure 2

Differences in plant development amongst ‘Type 1’, ‘Type 2’ and ‘Type 3’ seedlings in the Pyrus PEAR3 × ‘Moonglow’ progeny sown in Motueka in 2014. The letters on top of each box (a, b and c) represent significant differences (according to the Student–Newman–Keuls test). (a) Height of the seedlings measured at 30 (in light blue), 50 (in yellow) and 85 (in purple) days after germination. Significant differences amongst the three types are shown for each assessment. (b) Leaf area measured at 30 days after germination. (c) Average number of buds counted at 85 days after germination.

Figure 3

Genetic map of LG2 and LG5 of Pyrus PEAR3 and LG1 of ‘Moonglow’, indicating regions of segregation distortion. The MAF (red curves) is presented as a measure of segregation distortion of the markers evaluated on non-necrotic progeny. HRM and SSR markers used for ‘Type 1’ and ‘Type 2’ screening are highlighted in red. Newly mapped SNPs with respect to the map of Montanari et al are underlined. The regions involved in HN are marked in yellow. The locus let2 linked to ‘Type 2’ phenotype is in bold and italic.

Figure 4

Inheritance of the lethal alleles in the Pyrus PEAR3 × ‘Moonglow’ pedigree. Progenitors of PEAR3 and ‘Moonglow’ were scanned with SSR markers mapped within the regions involved in HN. For each marker, the incompatible allele (in bp) is highlighted in red.

Figure 5

Timing of the expression of the genetic incompatibilities and lethality that occur in the Pyrus PEAR3 × ‘Moonglow’ population. A timeline is drawn to show when ‘Type 1’ and ‘Type 2’ seedlings die or irreversibly stop growing and necrotize.